K R İ Z
Kriz Dergisi 9 ( 1 ) : 19-32
THE EPİDEMIOLOGY OF SUICIDE ATTEMPTS İN EUROPE+
Dr. Unni Bille-BRAHE*
Avrupa'da İntihar Girişimlerinin Epidemiyolojisi terinde endişe verici bir artış olduğunu bildirmekte
dir. Genel olarak en sık kullanılan intihar girişimi
ÖZET yöntemi ilaçla kendini zehirlemedir. Erkekler kadın
lara göre daha saldırgan intihar yöntemlerini tercih
Son 20-30 yıldır, ölümcül olmayan intihar dav etme eğilimindedirler.
ranışları bir çok üldeke büyüyen bir sorun olmuş
tur. Bir çok batı ülkesinde ölümle sonuçlanan inti Anahtar Kelimeler: İntihar, intihar girişimi, in
harlar 100 yıldan daha uzun bir süredir tihar davranışı, epidemiyoloji.
kaydedilmektedir. Ancak intihar girişimlerine ait
ulusal istatistikleri tutan hiçbir ülke yoktur. Bu ne SUMMARY
denle Dünya Sağlık Örgütü (DSÖ) Avrupa Bölge
sel Ofisi 1980'lerin ortalarında Avrupa ülkeleri ara During recent decades, non-fatal suicidal
sında intihar girişimleri ile ilgili bir çalışma behaviour has become a problem of increasing
başlatmıştır. "DSÖ/Avrupa Çok Merkezli intihar concern in mostcountries. Contrary to fatal suicidal
Davranışı Çalışması". Çalışmanın iki bölümü var acts that most westem countries have been
dır: izleme Çalışması ve Tekrarlanan intihar Giri- registering for more than 100 years, no country can
şimlerindeki Risk Faktörlerinin Belirlenmesi Çalış provide national statistics on suicide attempts. For
ması. İzleme Çalışması 1989'da başlatılmıştır. this reason, the European Regional Office of the
Başlangıçta çalışmaya 16 merkez katıldı. Ancak World Health Organisation initiated an
bunlardan bazıları bu konuya ayrılan ulusal kay inter-European study on suicide attempts in the
nakların yetersizliği nedeniyle çalışmadan ayrılmak mid-1980's. "The WHO/Euro Multicentre Study on
zorunda kaldı. Bununla beraber, yeni merkezler Suicidal Behaviour". There are two parts of the
çalışmaya katılmak istediler. Bugün için çalışmanın study: The Monitoring Part and The Repetition
İzleme Bölümü'nde 24 merkez yer almaktadır. Prediction Study. The Monitoring study started in
1989-1992 yılları arasında, 22.655 intihar girişimi 1989. Originally, 16 centres participated, butdue to
bildirilmiştir. Avrupa'da yaşa göre standardize edil lack of national funding some had to leave the
miş intihar girişimi hızı erkeklerde 100.000'de 136 study. On the other hand, new centres have
ve kadınlarda 100.000'de 186'dır. Erkeklerde en wanted to take part, and as of today, 24 centres
yüksek yaşa özel hız 25-34, kadınlarda ise 15-24 are included in the monitoring part of the project.
yaş gruplarında saptanmıştır. Bu dönem içinde er During the period of 1989-1992, 22.655 episodes
keklerin intihar hızlarında % 17, kadınların intihar of attempted suicide were reported, and
hızlarında ise % 14 düşüş olmuştur. Bununla birlik age-standardised average rates per 100.000 in
te bir çok merkez genç kadınlardaki intihar girişim- Europe were estimated to 136/100.000 for males
and 186/100.000 for females. The highest
age-specific rates for men were found among the
+ 36. Ulusal Psikiyatri Kongresi'nde Konferans 25-34 year-olds and for vvomen among the 15-24
konuşması olarak sunulmuştur. year-olds. There was a decrease during the period
Dr. Sosyolog WHO Collaborative Centre for
in the total rates of about 17% for men and 14% for
Prevention of Suicide Odense, Denmark.
K R İ Z
women. Most centres are, hovvever, expressing of risk factors and high-risk groups, and of
vvorries conceming markedly increasing characteristics of suicide attempters has so far
frequencies of suicide attempts among the young been based on results from various separate
girls. İn general the far most common method is studies wihch have been difficult to compare and to
self-poisoning, usually with overdoses of medicine. draw general conclusions from because of
There is a tendency tovvards men choosing more differences in chaice of method, case-finding
violent or determinant methods than vvomen. criteria, ete.
Key VVords: Suicide, suicide attempt, suicidal For these reasons, the European Regional
behaviour, epidemiology. Office of the VVorld Health Organization initiated an
inter-European study on suicide attempts in the
INTRODUCTION mid-1980's. The WHO/Euro Multicentre Study on
During recent decades, non-fatal suicidal Parasuicide (from 1998 on titled The WHO/Euro
behaviour has become a problem of increasing Multicentre Study on Suicidal Behaviour) was to
concem in most countries. Although there has cover two broad areas of research:
been a marked development in the medical * monitoring trends in the epidemiology of
treatment after a suicidal act and therefore greater parasuicide, ineluding identifying risk factors (the
probabilities of saving and survival, the frequency Monitoring Study); and
of non-fatal suicidal acts has been increasing vvith
an almost epidemic speed. İt is therefore argued * condueting follow-up studies of parasuicide
that the majority of these suicide attempts cannot populations as a special high-risk group for
be seen as 'unsuccessful suicides', but rather as a further suicidal behavior (the
special 'new' type of suicidal behaviour that is not Repetition-Prediction Study).
necessarily motivated by 'a wish to die', but which The main topic of this article will be the
is aiming at provoking changes in problematic or methods of and some results from the Monitoring
painful situations so that life (again) will be worth Study.
living. The Monitoring study started in 1989.
Hovvever, contrary to fatal suicidal acts that Originally, 16 centres participated, but due to lack
most vvestem countries have been registering for of national funding some had to leave the study.
more than 100 years, no country can provide On the other hand, new centres have wanted to
national statistics on suicide attempts, and our take part, and as of today, 24 centres are ineluded
knovvledge of the frequency of attempted suicide, in the monitoring part of the project.
Table 1. Status of centres participating in the WHO/Euro Multicentre Study on Suicidal Behaviour.
Old centres New Centres
Bern, Svvitzerland Mamak, Turkey
Helsinki, Finland Athens, Greece
Innsbruck Stadt & Land, Austria Cork, Ireland
Odense, Denmark Gent, Belgium
Oxford, UK Holon, Israel
Padova, Italy Limerick, Ireland
Sar-Trandelag, Norvvay Ljubljana, Slovenia
Stockholm, Svveden Pecs, Hungary
Umeâ, Svveden Rennes, France
VVûrzburg, Germany Tallinn, Estonia
Leiden, The Netherlands Riga, Latvia
No more participating Applied for parttetpatlon
Bordeaux, France Kiev, Ukraine
Cergy-Pontoise, France Odessa, Ukraine
Emilia Romagna, Italy Dresden City & VVeisseritz County, Germany
Novi Sad, Yugoslavia
K R İ Z
(T ' Riga
• Rennes Szeged ,
"Bern * ^ S & ^ ^ P « s
Bordeaux \ Emilia Romagna .
Fig. 1. Map of europe wİth participating centres.
K R İ Z
Initially, four main purposes were listed for about changes in the present situation. For a
this part of the project: discussion on the use of the terms parasuicide and
* to assess feasibility of using local case registers attempted suicide, see Bille-Brahe et al. 1994.
to monitör attempted suicide in defined Ali monitoring data are then transferred to the
catchment areas. The catchment areas are centre in VVürzburg, vvhere they are checked,
supposed to cover at least 200.000 inhabitants coordinated and re-checked by prof. Armin
(15yearsandover) Schmidtke and his co-workers.
* to estimate the true incidence of medically To provide an adequate epidemiological
treated suicide attempters and trends över time, description of the areas under study, each
using standardized definitions and case-finding participating centre has been required to submit
criteria standardized information about the catchment area
in question, including general background data
* to identify sociodemographic risk factors
such as the size and location of the area, the
significantly associated with attempted suicide
breakdovvn of the population according to age, sex
* to ascertain variations in patterns of treatment and civil status, housing, urban/rural areas and
following attempted suicide in different cultural unemployment rates, together with the main
contexts (with the aim of establishing more industrial and economic activities of the area.
effective services for preventing this type of These pieces of background information have
suicidal behaviour). been gathered at the centre in Odense and
Already in the planning phase did it become published in the booklet "Facts & Figures"
clear that for several reasons existing local case (Bille-Brahe (ed), 1999).
registers could not provide sufficiently reliable DATA ON ATTEMPTED SUİCİDE İN
data. A special monitoring of suicide attempts, EUROPE
using uniform monitoring forms and the same case Data from 15 centres for the first five-year
finding criteria, was therefore to be carried out at period (1989-1993) vvere published in 1994 in the
ali centres on consecutive episodes in ali places book Attempted Suicide in Europe (Kerkhof et al.
vvhere suicide attempters were seen and/or treated 1994) and in an article Attempted suicide in
by health personnel. A suicide attempt was defined Europe: rates, trends and sociodemographic
as: characteristics of suicide attempters during the
"an act with non-fatal outcome, in which an period 1989-1992 (Schmidtke et al. 1996). İn ali
individual deliberately initiates a non-habitual 22.655 episodes of attempted suicide vvere
behaviour that, if vvithout intervention from reported, and age-standardized average rates per
others, will cause self-harm, or deliberately 100.000 in Europe vvere estimated to 136/100.000
ingests a substance in excess of the for males and 186/100.000 for females. There was,
prescribed or generally recognized hovvever, huge differences betvveen the various
therapeutic dosage, and vvhich is aimed at areas under study: for men it varied from
realizing changes vvhich the subject desires 314/100.000 (in Helsinki) to 45/100.000 (in
via the actual or expected physical Guipüzcoa), and for vvomen from 462/100.000 (in
conseçuences". Cergy-Pontoise) to 69/100.000 (in Guipüzcoa). The
The definition includes acts that are highest age-specific rates for men vvere found
interrupted before the actual self-harm occurs (e.g. among the 25-34 year-olds and for vvomen among
if a person is removed from the railvvay track the 15-24 year-olds.
before the train arrives), but excludes acts of İn general, there was a decrease during the
self-harm by persons who do not understand the period in the total rates of about 17% for men and
meaning or the consequences of their act (e.g. 14% for vvomen. İt should be noted, hovvever, that
because of mental retardation or severe mental in four areas, namely Helsinki, Emilia Romagna,
illness). This operational definition does not specify Cergy-Pontoise, and VVürzburg total female rates
the strength of the wish to die, since this is almost had been increasing.
impossible to ascertain. That is ali intentional Table 2 gives an overvievv of the rates of
self-destructive behaviours are included, as long attempted suicide in the various part of Europe up
as these behaviours appear to be intended to bring to 1995
K R İ Z
«s CM •*> CD
2 S ^1 Z
< < û 2 UJ û
co o Z
U co a:
O O z n
uı i w
N SE Q
O CO B
LU D K o o û.
O (D o:
< co \-
LU CO O
? z *: <
15 g < ^ < LU O
LU Q O 7
_1 CO LU
co O LU u o
ne L3 Z O
o: LU 2 CO
z D (T <
LU T tü d oN LU UJ o z> l/J ü ->
OC V CO 3 Q N Q LU LU u 3
X. - 1 < t- N
O CO D.
Q LU > z
1- O CO
O o X et
z O z r)
LU CO a:
Fig. 2. Percentage changes in rates of attempted suicide (persons) by sex, 1989-1992 in 15 European areas under study
(Bern 1989-1990, Cergy-Pontoise and Guipüzcoa 1989-1991).
K R İ Z
Table 2. Rates of attempted suicide (events) per 100.000.15 years and older, in the areas under study. 1989-1995.
1989 1990 1991 1992 1993 1994 1995 ratio
M/F M/F M/F M/F M/F M/F M/F
Helsinki, Finland 330/237 340/266 323/247 314/238 270/214 333/267 272/232 1.18
Odense, Denmark 188/233 175/200 152/173 159/175 167/205 145/185 124/177 0.70
Stockholm, Sweden 179/314 176/227 115/192 148/195 147/224 124/190 106/162 0.66
Sar-Trandelag, Norway 147/210 145/210 151/177 142/169 113/163 118/112 92/117 0.79
Umeâ, Sweden 94/148 104/145 92/143 77/144 61/120 54/103 61/97 0.63
Bern, Switzerland 130/178 99/119 NA NA 105/130 74/111 60/117 0.51
Cork, Ireland* 196/206 0.96
Limerick, Ireland* 202/209 0.97
Dresden City, Germany*
VVeisseritz County, Germany*
InnsbruckStadt&Land.Austria 94/141 78/95 75/101 85/97 141/137 NA NA 1.03
Oxford, England 277/384 273/363 271/364 239/363 261/310 311/376 NA 0.85
Leiden, The Netherlands 81/148 102/144 82/129 78/134 NA NA NA 0.59
Rennes, France* 370/5401 380/5442
VVürzburg, Germany 72/100 66/84 68/105 55/108 101/174 92/142 77/127 0.61
Central Eastern Europe
Ljubljana, Slovenia* 84/79 0.95
Novi Sad, Yugoslavia*
Tallinn, Estonia* 278/188 1.48
Holon & Bat-Yam, Israel*
Padua, Italy 71/117 55/90 55/93 63/94 45/82 28/66 51/86 0.60
* Centres that just recently joined the VVHO/EURO Multicentre Study on Parasuicide.
1 The figures are for the town of Rennes.
2 The figures are for the Catchment area.
İn the publications mentioned above, an the two Irish areas, and in Ljubljana the ratios are
apparent increase in the sex ratio (m/f) was now getting close to 1.
discussed İt is interesting to note that in 1995, Fig. 3 shovvs the latest data available on the
there was an overvveight of men attempting suicide frequency of attempted suicide. Data on the
in three of the 15 areas under study, namely second five-year peiod 1993-1997 will be
Helsinki, Innsbruck, and Tallinn. Furthermore, in published shortly
K R İ Z
lOCOT-COlOlOCMCOOOCDCOCOCDCMCOCDlOinCM.- /rt co
O) O) O) O) O) O) O) Oî O) O) O) O O) O) 0>0> 05 0>0> O) <
0 z <
0^nZ3S ^ P Î30-JûuJO<SOü
Fıg. 3. Age-standardızed rates of attempted suicıde (events) per 100.000,15 years and older. Latest available year
K R İ Z
The figures indicate that the rank-order of the tendency tovvards men choosing more violent or
areas under study (cf. tab. 1) is rather stable, and determinant methods than women.
Males O Females
15-24 35-34 35-44 45-54 55-64 65-74 75-84 85+
Fig. 4. The age distribution of suicide atlempters, 1989-1993.
so is the distribution on age: in ali areas the SOCIODEMOGRAFHIC CHARACTERISTICS
highest rates are to be found among the young.
Most centres are, hovvever, expressing
vvorries conceming markedly increasing International literatüre is rich on studies on
risk factors and groups with increased risk for
frequencies of suicide attempts among the young
suicidal behaviour, and there seems to be a
girls. As yet, no pooled data on age-specific rates
general agreement as to the importance of at least
from the latest years are available, but as an
some parameters conceming demography, social
illustration of the problem, Danish rates of
conditions, living standard, health, and alcohol
attempted suicide for 15-19 year-olds and for
consumption and crimes. For this reason,
persons 20 years and older are shovvn in fig. 5. questions relating to such conditions were included
The figüre shows that while rates of attempted in the monitoring form, and descriptions of the
suicide in Denmark have göne down for ali age relevant parameters were included in the booklet
groups and for both sexes, during the 1990's the Facts & Figures. Results from the first period of the
15-19 year-old girls have almost tripled their rate. Monitoring Study confirm the existence of
The choice of method does differ betvveen the co-variations of certain sociodemographic and
areas, but in general the far most common method socioeconomic conditions, and the occurrence of
is self-poisoning, usually with overdoses of attempted suicide (Schmidtke et al. 1994) and it
medicine. İn particular, overdoses of various types was expected that there would be correlations
of mıld analgesics, especially Paracethamol between the rates of attempted suicide in the areas
agents, are very common among young girls. under study, and the said characteristics of the
İt can be seen from the figüre, that there is a areas.
K R İ Z
Rate per 100.000
— 15-19 dr
0 — ı 1 1 1 1 1 1 1
1990 1991 1992 1993 1994 1995 1996 1997
O -* ı 1 1 1 1 1 1 1
1990 1991 1992 1993 1994 1995 1996 1997
Fig. 5. Age and sex specific rates of attempted suicide (events) in Denmark, 1990-1997.
K R İ Z
80 B| Males O Females
Cutting Hanging Jumping Alcohol Moving obj. Firearms Submers. Car accıd.
Method of suicide attempt
Fig. 6. Methods of suicide attempts (average percentages).
However, as can be seen from table 3, While the risk of completed suicide is highest
correlations significant on the 95% level were among men and increasing with age, attempted
found regarding only two parameters, namely the suicides most often occur among young people
frequency of divorce, and the number of people and especially among young vvomen. Accordingly,
receiving public support (Bille-Brahe et al, 1996). the ratio suicide attempts/suicide differs -as shovvn
Somevvhat surprisingly, no correlations were found in fig. 8.- markedly vvith sex. That the variation is
betvveen e.g. unemployment or alcohol even bigger betvven age groups can be illustrated
consumption, and the frequency of suicide vvith another example from the Danish area under
attempts. study: here the ratio varies from 91:0 for 15-19
year-old girls to 1:1 for men 60 years and older.
This raises the question and the need for
further analyses: How then to explain the at times
rather marked differences betvveen the areas İt is, hovvever, interesting to note that the ratio
under study in the frequency of attempted suicide? attempted suicide/completed suicide, at least vvhen
ATTEMPTED SUİCİDE VERSUS it comes to mean or average rates, seems to be
COMPLETED SUİCİDE rather stable över the last many years. Considering
the fact that the population of attempters and of
From fig. 7 it is seen that the rate of suicides in many respects -as e.g. age and sex-
completed suicide varies markedly betvveen the differ significantly from each other (cf. Stengel,
areas under study, too. 1967), one may ask if there are some underlying
İn literatüre, it is often stated that the factors or conditions that in some way influence the
frequency of suicide attempts is probably 8-10 propensity to react to painful problems and
times higher than that of completed suicides. This sufferings vvith suicidal behaviour and that
might be true when mean total rates are knovvledge on such factors also may tribute to an
concerned, but in fact the statement is a gross explanation of the differences betvveen the centres
simplicism. discussed above.
K R İ Z
Table 3. Spearman's Rank correlations betvveen rates of parasuicide and some characteristics of the areas under study.
Correlation coeff. Significance
M F M F
Average population density in area (N-15) •0.0179 -0.0214 0.950 0.940
Urban/rural distribution (N=14) 0.4198 0.4901 0.135 0.075
Proportion vvorking in agriculture, forestry and fishery (N=12) 0.1160 0.2496 0.720 0.434
Sexratio(N=15) 0.1456 0.2695 0.605 0.331
Pet of the population above 40 years of age (N=15) 0.2687 0.1226 0.333 0.663
Number of people per houshold (N=9) •0.0924 0.0168 0.813 0.966
Single people in per cent of the total population (N=15) •0.1883 -0.3031 0.501 0.272
Single parent families in per cent of ali families (N=9) 0.5105 0.2762 0.160 0.472
Divorced in per cent of total population 0.8709 0.7593 0.000 0.004
Indicators ofUving Standard
Per capita income (N=11) 0.5455 0.4273 0.083 0.190
Percentage receiving public assistance (N=11) 0.8109 0.6834 0.002 0.020
Unempxloyment rate (N=12) -0.2557 -0.2557 0.422 0.422
Life expectancy (N=13) -0.3571 0.1209 0.231 0.694
Mortality rate per 100.000 per year (N=13) •0.3967 -0.3251 0.180 0.278
Infant mortality per 1000 live births (N=12) 0.1828 0.1511 0.570 0.639
Alcohol Consumption and Crime
Crimes per year per 1000 inhabitants (N=12) 0.4965 0.4755 0.101 0.118
Registered per capita alcohol consumption (N=15) •0.2594 -0.3399 0.351 0.215
Estimated per capita alcohol consumption (N=15) 0.0794 -0.0631 0.779 0.823
K R İ Z
CO o o.
_ O O. JZ
o 0) o •_
(0 .E~ o
—» C O C0*D >*<B C "O > » > » «
m Dİ Ec-
•- O ffl
i o s i«
* I a a 5 2-S
E o İS m
* 1a c
— j- O 3 S ö co co
•o 3 "=
c o. «
1 1 >• «0
E o x ~ C£ .
+ _ >.
I H İ İ .i l i
S "5 U
E o o
* £ « 1
o 3 l ^3
• ^ a »
o "5 2
X a »
.o- - T İS
= O s K
o 5 3 <D 'İS
5 5 =
z 1 w a. o
T- W (Ö"
Fig. 7. Rates of completed suicide per 100.000 inhab. 15 years and older in the areas under study, latest available data
(1991 to 1997).
K R İ Z
600 500 400 300 200 100 0 20 40 60 80 100
Attempted suıcide Completed suicide
600 500 400 300 200 100 0 20 40 60 80 100
Attempted sıucıde Completed sıncıde
Fıg 8 The ratıo attempted suıcıde/completed silicide by sex.
K R İ Z
CONCLUDING REMARKS taken place in the majority of the VVestern
European areas under study, but on the other hand
During later years, the rates of completed we find that young people and especially young
silicide have been decreasing in most VVestern vvomen are at an increasing risk. There is,
European countries, while the frequency has been hovvever, a lack of correlations betvveen known risk
increasing markedly in the Eastern and Central factors and the frequencies of attempted suicide,
European Countries (Bille-Brahe, 1998). İt is yet to and this -and also the apparent stability of the ratio
be seen vvhether this pattem repeats itself
attempted suicide/completed suicide- raise
regarding non-fatal suicidal acts but as more
questions that stili remain to be ansvvered. This
centres from the Eastern and Central part of
Europe join the WHO/Euro Multicentre Study, the calls for further analyses of our data and more
monitoring part of this project should be able to research on e.g. factors that can be related to the
answerthisquestion. societal structures in the various areas under
study, to pattems in social networks, and to
So far we can conclude that as far as attitudes tovvards suicidal behaviour, bound in
attempted suicide is concerned, a decrease has cultural and religious traditions.
REFERENCES Findings from the Multicentre Study on Parasuicide by
the WHO Regional Office for Europe. DSWO Press,
Bille-Brahe U. Suicidal Behaviour in Europe. The Leiden, 1994.
situation in the 1990s. WHO, Copenhagen, 1998.
Schmidtke A, Bille-Brahe U, De Leo D, Kerkhof A,
Bille-Brahe U (ed). Facts and Rgures. 2nd. ed. Bjerke T et al. Attempted suicide in Europe: rates, trends
WHO, Copenhagen, 1999. and sociodemoegraphic characteristics of suicide
attempters during the period 1989-1992. Results from
Bille-Brahe U, Andersen K, VVasserman D, the WHO/Euro Multicentre Study on Parasuicide. Açta
Schmidtke A, Bjerke T et al. The WHO/Euro Multicentre Psychiatr Scand 1996,93:327-38.
Study: Risk of Parasuicide and the comparability of the
areas under study. CRISIS 1996,17/1:32-42. Schmidtke A, Bille-Brahe U, Kerkhof A, De Leo D,
Bjerke T et al. Sociodemographic Characteristics of
Bille-Brahe U, Schmidtke A, Kerkhof AJMF, De Leo suicide attempters in Europe. İn: Kerkhof AJFM,
D, Lönnqvist J, Platt S. Background and the Introduction Schmidtke A, Bille-Brahe U, De Leo D, Lönnqvist J
to the study. İn: Kerkhof et al (eds.) Attempted suicide in (eds). Attempted Suicide in Europe. Findings from the
Europe. Findings from the Multicentre Study on Multicentre Study on Parasuicide by the WHO Regional
Parasuicide by the WHO Regional Office for Europe. Office for Europe. Chapter 14, pp. 231-241. DSWO
Chapter 1. pp. 11-15. DSWO Press, Leiden, 1994. Press, Leiden, 1994.
Kerkhof AJMF, Schmidtke A, Bille-Brahe U, De Leo Stengel E. Suicide and attempted suicide. Penguin
D, Lönnqvist J. (eds). Attempted Suicide in Europe. Books, London, 1967.